duong tam kien

Where until now, node-link layouts have emphasized representing relationships and have too often been an end of a process (feed in the data, see what it looks like, go back to the data), the abliity to directly access data, write queries, and closely investigate individual nodes show the potential of how visualizations can become part of the analysis process itself, and to answer questions such as "Why do some nodes cluster together?" "Which nodes in the entire graph share the same attributes?"

The distinction between visualization and interface is blurring, and the line between visualization and analysis is being crossed.

Getting and Understanding the Bigger Picture